Gambler's fallacy
The gambler’s fallacy is the mistaken belief that a random outcome becomes more likely because it has not happened for a while, or less likely because it has just happened several times in a row. On a fair roulette wheel or a certified slot, each result is independent of the last, so a run of blacks makes red no more “due” than it was before. The error is also called the Monte Carlo fallacy, after a famous night at the roulette tables in 1913.
Why independent results have no memory
The whole fallacy rests on one idea, independence, which means the probability of the next result is the same whatever came before it. A European roulette wheel has 37 pockets, 18 of them black, so black lands with a probability of 18 in 37, or about 48.65%, on every spin.
That figure does not move after a streak.
Where readers get tangled is the difference between two questions. Before any spin, the chance of seeing black 26 times running is 18/37 multiplied by itself 26 times, which works out at roughly 1 in 136.8 million. Once 25 blacks have already landed, though, those spins are history, and the chance that the 26th is black is simply 18 in 37 again. The rare sequence was rare before it started; the single next spin never is.
The Monte Carlo night
On 18 August 1913, at the Casino de Monte-Carlo, the ball fell on black 26 times in a row, and players lost millions of francs pushing money onto red as the run lengthened. Accounts of the evening have been retold for over a century and the sums vary between them, but the reasoning error behind it is documented consistently.
Gambler's fallacy versus the hot hand
The gambler’s fallacy is most often confused with its mirror image, the hot-hand fallacy, which expects a streak to continue rather than end. The rows below set the two side by side; where both beliefs share the same flaw, the row is marked as such.
Gambler's fallacy | Hot-hand fallacy |
|---|---|
| Expects the streak to reverse | Expects the streak to continue |
| After ten blacks, bets on red | After ten blacks, bets on black |
| Often feeds stake escalation such as doubling after losses | Often feeds raising stakes while winning |
| Ignores that each spin is independent | Ignores that each spin is independent |
Where players meet it on slots and tables
On slots, the fallacy usually shows up as the belief that a machine which has not paid for a while is “due” a bonus round. An online slot draws each spin from a random number generator, and the RNG is tested by independent labs precisely so that one spin cannot inform the next. A long dry spell tells the player nothing about the coming spin.
RTP is where the misunderstanding gets subtle, and it is worth taking slowly. A slot with a published RTP of 96% is expected to return 96 cent of every euro staked over a very large number of spins, across all players, so €500 of wagering carries a long-run expected cost of about €20. That is an average, not a balance the game owes anyone: a session that has run below 96% will not be pushed back up to it, and studios often release the same title in several RTP versions, so the figure depends on which version the operator runs.
Long-run averages are reached by dilution, not repayment.
The same logic applies at the table. Live roulette, baccarat shoes and dice games all produce streaks, and trend boards beside the wheel make those streaks very visible, but a board showing past results changes nothing about the next one.
Common confusions and the one real exception
The law of large numbers is the usual source of the mix-up. It says the proportion of reds and blacks drifts towards the true odds as spins pile up, and readers often take that to mean a shortfall gets corrected. It does not: 26 extra blacks stay 26 extra blacks, they simply matter less as a share of a million spins.
There is one genuine case where past results change future odds, and that is when cards are dealt without replacement. In blackjack, a card that has left the shoe cannot reappear until the shuffle, so the composition of the remaining cards really does shift, which is the basis of card counting. That is dependence, a different situation entirely, and it does not carry over to roulette spins, dice or RNG slots.
What we check first, when a reader describes a system built on streaks, is whether the game actually has memory.
